U.S. patent application number 13/107368 was filed with the patent office on 2011-11-17 for x-ray computed tomography apparatus.
Invention is credited to Satoru NAKANISHI, Yasuhiro NOSHI.
Application Number | 20110280459 13/107368 |
Document ID | / |
Family ID | 44911804 |
Filed Date | 2011-11-17 |
United States Patent
Application |
20110280459 |
Kind Code |
A1 |
NAKANISHI; Satoru ; et
al. |
November 17, 2011 |
X-RAY COMPUTED TOMOGRAPHY APPARATUS
Abstract
According to one embodiment, an X-ray computed tomography
apparatus includes an X-ray tube, an X-ray detector, and a rotating
unit. The first reconstruction processing reconstructs a clinical
image based on projection data detected by the X-ray detector. The
second reconstruction processing reconstructs a noise image based
on noise data. The clinical image is combined with the noise
image.
Inventors: |
NAKANISHI; Satoru;
(Utsunomiya-shi, JP) ; NOSHI; Yasuhiro;
(Otawara-shi, JP) |
Family ID: |
44911804 |
Appl. No.: |
13/107368 |
Filed: |
May 13, 2011 |
Current U.S.
Class: |
382/131 ;
378/4 |
Current CPC
Class: |
G06T 11/006 20130101;
A61B 6/032 20130101; A61B 6/5258 20130101; A61B 6/583 20130101 |
Class at
Publication: |
382/131 ;
378/4 |
International
Class: |
G06K 9/00 20060101
G06K009/00; A61B 6/03 20060101 A61B006/03 |
Foreign Application Data
Date |
Code |
Application Number |
May 17, 2010 |
JP |
2010-113521 |
Claims
1. An X-ray computed tomography apparatus, comprising: an X-ray
tube configured to generate an X-ray; an X-ray detector configured
to generate projection data by detecting the X-ray transmitted
through an object; a rotating unit configured to rotatably support
both the X-ray tube and the X-ray detector around the object; a
first reconstruction processing unit configured to reconstruct a
clinical image by first reconstruction processing based on the
generated projection data; a second reconstruction processing unit
configured to reconstruct a noise image by second reconstruction
processing different from the first reconstruction processing based
on noise data; and an image combining processing unit configured to
combine the clinical image with the noise image.
2. The apparatus of claim 1, wherein noise reduction effect of the
second reconstruction processing is smaller than that of the first
reconstruction processing, and image granularity of the second
reconstruction processing is greater than that of the first
reconstruction processing.
3. The apparatus of claim 1, wherein the second reconstruction
processing comprises one of filtered back projection processing and
other type of analytical reconstruction processing.
4. The apparatus of claim 3, wherein a reconstruction function
selected from a plurality of reconstruction functions in accordance
with an operator instruction is applied.
5. The apparatus of claim 1, wherein the first reconstruction
processing comprises one of ART processing and other type of
iterative approximation processing.
6. The apparatus of claim 1, wherein the first reconstruction
processing comprises ART processing, and the second reconstruction
processing comprises filtered back projection processing.
7. The apparatus of claim 1, wherein the first reconstruction
processing reduces noise of the clinical image.
8. The apparatus of claim 1, wherein the noise data corresponds to
Gaussian noise with an average value of 0 and a standard deviation
.sigma..
9. The apparatus of claim 1, wherein the noise data comprises
projection data associated with a plurality of views having only
noise components and obtained by scanning a phantom made of a
homogeneous material.
10. The apparatus of claim 1, wherein the clinical image and the
noise image comprise volume data.
11. The apparatus of claim 1, further comprising a noise image
processing unit configured to apply enlargement/reduction
processing to the noise image in accordance with a pixel size of
the clinical image and convolute one of a real spatial distribution
and a frequency spatial distribution corresponding to an arbitrary
reconstruction function with respect to a noise image to which the
enlargement/reduction processing is applied.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is based upon and claims the benefit of
priority from Japanese Patent Application No. 2010-113521, filed
May 17, 2010; the entire contents of which are incorporated herein
by reference.
FIELD
[0002] Embodiments described herein relate generally to an X-ray
computed tomography apparatus.
BACKGROUND
[0003] Image noise reduction processing changes the granularity of
image noise. Noise reduction processing is effective for the image
reconstructed by iterative approximation processing. On the other
hand, the granularity greatly deteriorates. Many observers have
long experience in observing images having certain
granularities.
[0004] The prior art includes a method of combining a
"reconstructed image having undergone noise reduction by iterative
approximation processing" with an "original image". Combining the
original image will add granularity components to the image having
undergone noise reduction processing, thereby minimizing a sense of
discomfort in appearance.
[0005] A problem of this technique is that the artifact components
of the original image are also added to the reconstructed image to
result in a reduction in image improving effect. When obtaining
granularity like that of an image reconstructed by filtered back
projection processing (to be referred to as an FBP image
hereinafter) from a "reconstructed image having undergone noise
reduction by iterative approximation processing", if the input
source data is minority data, a deterioration in image quality
becomes noticeable because aliasing artifact is noticeable in FBP
reconstruction for the reconstruction of an original image.
[0006] Another problem in the prior art is that although FBP
reconstruction can control granularity by controlling the frequency
characteristics of a ramp filter called a reconstruction function,
it is theoretically difficult to provide a unit for operating
granularity for an image reconstructed by iterative approximation
processing. This makes it difficult to obtain granularity in
accordance with the preferences of customers.
BRIEF DESCRIPTION OF THE DRAWINGS
[0007] FIG. 1 is a view showing the arrangement of an X-ray
computed tomography apparatus according to the first
embodiment;
[0008] FIG. 2 is a flowchart showing a procedure for generating a
final image (volume) according to the first embodiment;
[0009] FIG. 3 is a view showing the arrangement of an X-ray
computed tomography apparatus according to the second embodiment;
and
[0010] FIG. 4 is a flowchart showing a procedure for generating a
final image (volume) according to the second embodiment.
DETAILED DESCRIPTION
[0011] In general, according to one embodiment, an X-ray computed
tomography apparatus includes an X-ray tube, an X-ray detector, and
a rotating unit. The first reconstruction processing reconstructs a
clinical image based on projection data detected by the X-ray
detector. The second reconstruction processing reconstructs a noise
image based on noise data stored in advance. The clinical image is
combined with the noise image and displayed (frame addition
processing).
[0012] The X-ray computed tomography apparatus according to this
embodiment will be described below with reference to the views of
the accompanying drawing.
[0013] First of all, a basic feature of this embodiment is to
provide granularity for a clinical image by combining a
"reconstructed image (clinical image) having undergone noise
reduction by iterative approximation processing" with an "image
(noise image) having noise components". A "noise image" to be
combined with a clinical image is stored upon being reconstructed
in advance by filtered back projection processing using a specific
reconstruction function or another type of analytical
reconstruction processing from projection data (e.g., projection
data associated with uniform noise) associated with a plurality of
views having only noise components (first embodiment).
Alternatively, it is possible to individually reconstruct a "noise
image" to be combined with a clinical image by storing projection
data associated with a plurality of views having only noise
components in advance and performing filtered back projection
processing using a reconstruction function arbitrarily selected by
the operator or another type of analytical reconstruction
processing at the same resolution as that set in scanning for the
acquisition of projection data associated with an object which are
used for clinical image reconstruction (second embodiment).
[0014] Note that X-ray computed tomography apparatuses include a
rotate/rotate-type apparatus in which an X-ray tube and a radiation
detector rotate together around an object to be examined and a
stationary/rotate-type apparatus in which many detection elements
are arrayed in the form of a ring, and only an X-ray tube rotates
around an object. This embodiment can be applied to either type. In
this case, the rotate/rotate type, which is currently the
mainstream, will be exemplified. In order to reconstruct one-slice
tomogram data, projection data corresponding to one rotation around
an object, i.e., about 360.degree., is required, or
180.degree..+-..alpha. (.alpha.: fan angle) projection data is
required in the half scan method. The embodiment can be applied to
either of these reconstruction schemes. As units of converting
incident X-rays into electric charges, the following techniques are
the mainstream: an indirect conversion type that converts X-rays
into light through a phosphor such as a scintillator and converts
the light into electric charges through photoelectric conversion
elements such as photodiodes, and a direct conversion type that
uses generation of electron-hole pairs in a semiconductor by X-rays
and migration of the electron-hole pairs to an electrode, i.e., a
photoconductive phenomenon. As an X-ray detection element, either
of these schemes can be used. In this case, the former type, i.e.,
the indirect conversion type, will be exemplified. Recently, with
advances toward the commercialization of a so-called multi-tube
type X-ray computed tomography apparatus having a plurality of
pairs of X-ray tubes and X-ray detectors mounted on a rotating
ring, related techniques have been developed. The embodiment can be
applied to both a conventional single-tube type X-ray computed
tomography apparatus and a multi-tube type X-ray computed
tomography apparatus. The single-tube type X-ray computed
tomography apparatus will be exemplified here.
[0015] FIG. 1 shows the arrangement of the X-ray computed
tomography apparatus according to the first embodiment. The X-ray
computed tomography apparatus according to the first embodiment
includes a host controller 110 as a main control unit and a gantry
100. An opening portion is formed in the central portion of the
gantry 100. An object is placed on the top of a bed (not shown) and
inserted into the opening portion. The gantry 100 includes an X-ray
tube 101 and an X-ray detector 103. As the X-ray tube 101, an X-ray
tube of a type that generates an X-ray cone beam is used in
accordance with cone beam scanning. As the X-ray detector 103, a
multi-slice type or two-dimensional array type detector having
X-ray detection elements arrayed two-dimensionally is used. The
X-ray detector 103 is mounted on a ring-like rotating frame 102
which is rotated by a gantry driving unit 107, together with the
X-ray tube 101.
[0016] A high voltage generator 109 applies a tube voltage (high
voltage) between the cathode and anode of the X-ray tube 101
through a slip ring 108. The high voltage generator 109 also
supplies a filament current to the filament of the X-ray tube 101.
The application of a tube voltage and the supply of a filament
current will generate X-rays The X-ray detector 103 includes a
plurality of X-ray detection elements each having, for example, a
0.5 mm.times.0.5 mm square light-receiving surface. For example,
916 X-ray detection elements are arrayed in an arcuated form along
the channel direction. For example, 64 detection element rows are
arranged parallel in the slice direction.
[0017] A data acquisition circuit 104 generally called a DAS (Data
Acquisition System) converts a signal output from the X-ray
detector 103 for each channel into a voltage signal, amplifies it,
and further converts it into a digital signal. This data (to be
also referred to as raw data) is sent to a preprocessing device 106
via a noncontact data transmitter 105 using magnetism or light as a
medium. The preprocessing device 106 performs correction processing
such as sensitivity correction for the raw data. The preprocessed
raw data is generally called projection data. A projection data
storage unit 112 stores each projection data in association with a
view angle representing the rotational angle of the X-ray tube 101,
a channel number, a column number, and a code representing the
position of the top. Note that the projection data actually
acquired by scanning an object will be referred to as clinical
projection data to discriminate it from projection data associated
with only noise components (to be described later). In addition,
projection data associated with only noise components will be
referred to as noise projection data.
[0018] A clinical image generation processing unit 118 generates
clinical image data based on clinical projection data. Clinical
image data is a secondary original image (slice image) or a
tertiary original image (volume). Clinical image data is a
low-noise image, as indicated by "S1" in FIG. 2. Image generation
processing performed by the clinical image generation processing
unit 118 typically corresponds to ART (Arithmetic Reconstruction
Processing), a MAP-EM method, or another type of iterative
approximation processing. Image reconstruction processing may be
processing corresponding to filtered back projection processing, a
convolution integral method, a Fourier transform method, or another
analytical reconstruction method. When using processing
corresponding to filtered back projection processing or another
analytical reconstruction method, noise reduction filtering
processing is combined with processing corresponding to this
analytical reconstruction method. This generates a low-noise
clinical image. A typical example of noise reduction filtering
processing is spatial smoothing processing.
[0019] A noise image storage unit 119 stores the data of a noise
image representing granularity higher than that of a clinical
image. A noise image is reconstructed in advance with a specific
spatial resolution (a specific pixel size) by filtered back
projection processing or another type of analytical reconstruction
processing using a specific reconstruction function based on
projection data associated with a plurality of views having only
noise components obtained by scanning, for example, a phantom made
of a homogeneous material. The noise image storage unit 119 then
stores the noise image. A noise component is defined by Gaussian
noise with an average value of 0 and a standard deviation
.sigma..
[0020] A noise image processing unit 120 applies
enlargement/reduction processing, convolution processing (S2 in
FIG. 2), and spatial smoothing processing to a noise image.
Enlargement/reduction processing is processing for equalizing the
spatial resolution (pixel size) of the noise image to the spatial
resolution (pixel size) of the clinical image.
[0021] Convolution processing is convolution of a spatial
distribution corresponding to a reconstruction function in a real
space or frequency space for a noise image. The spatial
distribution corresponding to the reconstruction function is a
granularity distribution obtained by reconstructing uniform
projection data by processing corresponding to filtered back
projection processing. A plurality of granularity distributions are
generated in advance by using a plurality of kinds of
reconstruction functions and stored in a reconstruction function
distribution storage unit 121. The operator can arbitrarily select
a granularity distribution via an input device 115. Selecting a
granularity distribution makes it possible to arbitrarily control
the granularity of a noise image.
[0022] Adjusting a smoothing coefficient for smoothing processing
can arbitrarily control noise intensity. When adjusting a smoothing
coefficient, the operator may arbitrarily select or designate a
smoothing coefficient via the input device 115. Alternatively, it
is possible to estimate the intensity of image noise from a
clinical image, typically calculate a standard deviation, and
select one of smoothing coefficients, associated with standard
deviations in advance, in accordance with the calculated standard
deviation.
[0023] An image combining processing unit 122 combines a low-noise
clinical image with a noise image having undergone image processing
by a noise image processing unit 120 (S3 in FIG. 2), thereby
obtaining a final image with granularity being added to the
clinical image. This combining processing is typically weighted
addition of pixels. A uniform weight may be provided for an entire
image area. Alternatively, it is possible to decrease the weight
for a noise image with respect to an edge portion (an area where
the spatial frequency is relatively high) of a region of a clinical
image so as to relatively increase the contribution ratio of the
clinical image and suppress noise. In contrast, it is possible to
increase the weight for the noise image with respect to a uniform
portion (an area where the spatial frequency is relatively low)
exhibiting small changes in the luminance of the clinical image so
as to decrease the contribution ratio of the clinical image and
enhance noise. A display device 116 displays the final image.
[0024] As described above, it is possible to provide granularity
for a clinical image by combining a "reconstructed image (clinical
image) having undergone noise reduction by iterative approximation
processing or the like" with an "image (noise image) having noise
components". In addition, selecting a granularity distribution can
arbitrarily adjust the degree of granularity.
[0025] FIG. 3 shows the arrangement of an X-ray computed tomography
apparatus according to the second embodiment. The same reference
numerals as in FIG. 3 denote the same parts in FIG. 1, and a
description of them will be omitted.
[0026] In the first embodiment described above, a noise image
prepared in advance is properly processed and combined with a
clinical image. In the second embodiment, projection data (to be
referred to as noise projection data) covering a plurality of views
having only noise components is prepared in advance. A noise image
is then reconstructed from this noise projection data by filtered
back projection processing, a convolution integral method, a
Fourier transform method, or another analytical reconstruction
method, and combined with a clinical image.
[0027] A noise projection data storage unit 123 stores projection
data noise projection data associated with a plurality of views
having only noise components. This noise projection data is formed
by a Gaussian noise model with an average value of 0 and a standard
deviation .sigma. or another kind of noise model. Alternatively,
noise projection data is obtained by scanning, for example, a
phantom made of a homogenous material. The format of this noise
projection data may differ from that of real projection data. For
example, if the data count of projection data actually acquired
from an object is 100 views/rotation, the data count of noise
projection data may be 100 views/rotation or less, e.g., 80
views/rotation, or more, e.g., 120 views/rotation. Typically, the
data count of noise projection data is preferably equal or
approximate to that of projection data actually acquired from an
object so as not to increase the sense of discomfort of
granularity.
[0028] A noise image generation processing unit 125 generates a
noise image by processing corresponding to processing corresponding
to filtered back projection processing, a convolution integral
method, a Fourier transform method, or another analytical
reconstruction method using an arbitrary reconstruction function
based on noise projection data (S4 in FIG. 4). A noise image is
reconstructed in advance at the same spatial resolution (pixel
size) as that in reconstruction processing of a clinical image. The
operator arbitrarily selects a reconstruction function used by the
noise image generation processing unit 125 from a plurality of
reconstruction functions stored in a reconstruction function
storage unit 124 via an input device 115. Selecting this
reconstruction function makes it possible to arbitrarily adjust the
granularity of a noise image. The reconstruction function to be
selected should typically differ from the reconstruction function
used by a clinical image generation processing unit 118 when it
uses an analytical reconstruction method.
[0029] An image combining processing unit 122 can obtain a final
image with granularity being added to a low-noise clinical image by
combining the clinical image with the noise image generated by the
noise image generation processing unit 125.
[0030] As described above, it is possible to provide granularity
for a clinical image by combining a "reconstructed image (clinical
image) having undergone noise reduction by iterative approximation
processing or the like" with an "image (noise image) having noise
components", and to arbitrarily adjust the degree of granularity by
selecting a reconstruction function used in noise image
reconstruction.
[0031] While certain embodiments have been described, these
embodiments have been presented by way of example only, and are not
intended to limit the scope of the inventions. Indeed, the novel
embodiments described herein may be embodied in a variety of other
forms; furthermore, various omissions, substitutions and changes in
the form of the embodiments described herein may be made without
departing from the spirit of the inventions. The accompanying
claims and their equivalents are intended to cover such forms or
modifications as would fall within the scope and spirit of the
inventions.
* * * * *